基于色彩衰减补偿和Retinex的水下图像增强
PDF下载 (235)杨福豪,史启超,蓝方鸣,彭宗举.基于色彩衰减补偿和Retinex的水下图像增强[J].宁波大学学报(理工版),2020,33(1):58-64.DOI:
YANG Fuhao,SHI Qichao,LAN Fangming,PENG Zongju.Underwater image enhancement based on color attenuation compensation and Retinex[J].Journal of Ningbo University(Natural Science & Engineering Edition),2020,33(1):58-64.DOI:
| Title: | Underwater image enhancement based on color attenuation compensation and Retinex |
| 作者: | 杨福豪, 史启超, 蓝方鸣, 彭宗举 |
| Author(s): | YANG Fuhao, SHI Qichao, LAN Fangming, PENG Zongju |
| 关键词: | 水下图像增强; 色彩补偿; 导向滤波; 对比度 |
| Keywords: | underwater image enhancement; color compensation; guided filtering; contrast |
| 分类号: | TP391 |
| 文献标识码: | A |
| 摘要: | 针对水下图像存在的色偏、雾状模糊、低曝光和非均匀光照问题, 提出一种基于色彩衰减补偿和Retinex的水下图像增强算法. 首先, 为校正水下图像的色偏, 利用水体对不同波长光线衰减不一致的特性自适应地补偿其R、G、B通道. 然后, 使用基于多尺度导向滤波的Retinex去除雾状模糊, 增强对比度. 最后, 根据水下图像和自然图像的直方图分布特征对其进行归一化处理, 从而在保存图像主要信息的前提下增强其纹理和曝光. 实验结果表明, 该算法不但具有较优的视觉感知效果, 而且具有较高的图像质量评价分数. 该算法具有较强的适应性, 有助于计算机视觉算法在水下的应用. |
| Abstract: | To tackle the problems of color cast, foggy blur, low exposure and non-uniform illumination in underwater images, an underwater image enhancement algorithm is proposed based on attenuation compensation of color and the theory of Retinex. Firstly, in order to correct the color cast in underwater images, the red, green, and blue channel are adaptively compensated, which is achieved by using the different wavelengths of light attenuated accordingly. Then, the Retinex based on multi-scale guided filtering is used to remove foggy blur and improve contrast. Finally, the deblurred underwater image is normalized according to histogram distribution of the underwater image and the natural image, which enhances its texture and exposure on the premise of retaining the main information of images. The experimental results show that the proposed algorithm not only offers the best visual perception, but also achieves the highest score of image quality evaluation. The proposed algorithm features high adaptability, and may provide references for research in underwater target identification. |
| 参考文献 /References: | [1] Wang Y, Song W, Fortino G, et al. An experimental-based review of image enhancement and image restoration methods for underwater imaging[J]. IEEE Access, 2019, 7:140233-140251. [2] Amer K O, Elbouz M, Alfalou A, et al. Enhancing underwater optical imaging by using a low-pass polarization filter[J]. Optics Express, 2019, 27(2):621-643. [3] 吕文磊, 张旭, 刘可. 一种距离选通水下激光成像系统设计与实验研究[J]. 兵器装备工程学报, 2019, 40(8):198-202. [4] 黄冬梅, 王龑, 宋巍, 等. 不同颜色模型下自适应直方图拉伸的水下图像增强[J]. 中国图象图形学报, 2018, 23(5):640-651. [5] Zhang W, Li G, Ying Z. A new underwater image enhancing method via color correction and illumination adjustment[C]. 2017 IEEE Visual Communications and Image Processing, 2017:1-4. [6] Ghani A S A. Image contrast enhancement using an integration of recursive-overlapped contrast limited adaptive histogram specification and dual-image wavelet fusion for the high visibility of deep underwater image[J]. Ocean Engineering, 2018, 162:224-238. [7] Drews P L J, Nascimento E R, Botelho S S C, et al. Underwater depth estimation and image restoration based on single images[J]. IEEE Computer Graphics and Applications, 2016, 36(2):24-35. [8] 王鑫, 朱行成, 宁晨, 等. 融合暗原色先验和稀疏表示的水下图像复原[J]. 电子与信息学报, 2018, 40(2):264-271. [9] Yang M, Sowmya A, Wei Z Q, et al. Offshore underwater image restoration using reflection-decomposition-based transmission map estimation[J/OL]. IEEE Journal of Oceanic Engineering [2019-03-17]. https://ieeexplore.ieee.org/document/8605308. [10] Li C, Quo J, Pang Y, et al. Single underwater image restoration by blue-green channels dehazing and red channel correction[C]. IEEE International Conference on Acoustics, 2016. [11] Lu J, Li N, Zhang S, et al. Multi-scale adversarial network for underwater image restoration[J]. Optics & Laser Technology, 2019, 110:105-113. [12] Chen X, Yu J, Kong S, et al. Towards real-time advancement of underwater visual quality with GAN[J]. IEEE Transactions on Industrial Electronics, 2019, 66(12):9350-9359. [13] Cheng N, Zhao T, Chen Z, et al. Enhancement of underwater images by super-resolution generative adversarial networks[C]. Proceedings of the 10th International Conference on Internet Multimedia Computing and Service, ACM, 2018:22. [14] Land E H. The Retinex theory of color vision[J]. Scientific American, 1977, 237(6):108-129. [15] Rahman Z U, Woodell G A. Multi-scale Retinex for color image enhancement[C]. International Conference on Image Processing, IEEE, 2002. [16] He K, Sun J, Tang X. Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 35(6):1397-1409. [17] Eustice R, Pizarro O, Singh H, et al. UWIT: Underwater image toolbox for optical image processing and mosaicking in MATLAB[C]. Proceedings of the 2002 International Symposium on Underwater Technology, IEEE, 2002:141-145. [18] Buchsbaum G. A spatial processor model for object colour perception[J]. Journal of the Franklin Institute, 1980, 310(1):1-26. [19] van de Weijer J, Gevers T, Gijsenij A. Edge-based color constancy[J]. IEEE Transactions on Image Processing, 2007, 16(9):2207-2214. [20] Zhu Q, Mai J, Shao L. A fast single image haze removal algorithm using color attenuation prior[J]. IEEE Transactions on Image Processing, 2015, 24(11):3522-3533. [21] Yang M, Sowmya A. An underwater color image quality evaluation metric[J]. IEEE Transactions on Image Processing, 2015, 24(12):6062-6071. [22] Panetta K, Gao C, Agaian S. Human-visual-system-inspired underwater image quality measures[J]. IEEE Journal of Oceanic Engineering, 2015, 41(3):541-551. [23] Wang S, Ma K, Yeganeh H, et al. A patch-structure representation method for quality assessment of contrast changed images[J]. IEEE Signal Processing Letters, 2015, 22(12):2387-2390. |
| 备注/Memo: | 收稿日期: 2019－07－06. 宁波大学学报（理工版）网址: http://journallg.nbu.edu.cn/ 基金项目: 国家自然科学基金（61771269）; 浙江省自然科学基金（LY17F010005）. 第一作者: 杨福豪（1992－）, 男, 河南汝南人, 在读硕士研究生, 主要研究方向: 水下图像处理. E-mail: 306733695@qq.com *通信作者: 彭宗举（1973－）, 男, 四川南部人, 教授, 主要研究方向: 视频与图像信号处理. E-mail: pengzongju@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
